@jacobli99: If we are ever to build machines that can operate in new domains like experts, either we must reduce each domain to a s…
Summary
Jacob X. Li discusses the need for AI systems to develop expertise autonomously from a corpus of documents, framing this as a challenging form of continual learning.
View Cached Full Text
Cached at: 06/18/26, 02:02 AM
If we are ever to build machines that can operate in new domains like experts, either we must reduce each domain to a sufficiently verifiable environment, or we must develop machines that can study to gain expertise on their own. Machine Studying explores the second possibility!
Another instance of the “intelligence = compression” intuition. If self-directed study ahead of time can improve performance over unlimited test-time lookup, most likely it’s because the agent has found a way to generalize and compress the raw facts into higher-level denser “learnings”. Improvement on this axis will be a very interesting canary in the coal mine that RSI is starting to work.
Nice work @jacobli99, @lateinteraction, Rick and team!
Similar Articles
@lateinteraction: Been extremely excited about this work by @jacobli99! We're disappointed in the current ways our agents develop experti…
A discussion on the challenge of enabling AI systems to develop deep expertise from documents, akin to humans learning from textbooks, highlighting a form of continual learning.
@jacobli99: Continual learning is widely discussed right now, but mostly as improving on the job or avoiding catastrophic forgettin…
Jacob Li introduces the concept of 'Machine Studying' as a distinct and urgent form of continual learning, where AI systems must develop expertise in a new domain from a corpus of documents alone.
@jacobli99: To compare procedures for machine studying, we start by defining expertise. The corpus is always available at test time…
Jacob Li introduces 'Machine Studying' as a new problem in continual learning: how AI systems develop expertise in an unfamiliar domain given only a corpus of documents, distinct from avoiding catastrophic forgetting.
@lateinteraction: one of the least expected side effects of this effort is that it gives us a measurable definition of agent “intelligenc…
Jacob X. Li discusses a new perspective on continual learning for AI, emphasizing developing expertise from a corpus of documents, and suggests it provides a measurable definition of agent intelligence.
@DSPyOSS: “In a way, Machine Learning asks how a system can improve from data when we have a precise objective to optimize. Machi…
A twitter thread by @DSPyOSS and Jacob X. Li contrasts Machine Learning (optimizing from data with a precise objective) with 'Machine Studying' (learning from a declarative corpus without a downstream task), highlighting the urgent need for AI systems to develop expertise from unstructured documents.